8 Oct
2005
8 Oct
'05
7:07 p.m.
As the subject implies, I've got a PCjr with a floppy drive that is having
trouble reading disks and fails the internal CTRL-ALT-INS diagnostics.
Symptoms include a very loud/bad noise at first seek (rest of seeks sound ok),
and a "B" ("have drive serviced") result of the diags. The diskette I
used for
diags was tested in a 5150 and found to be good. This drive was working a week
ago but has declined steadily until it can't successfully boot the machine. (I
bought some time by formatting disks 8 sectors per track instead of 9, but that
no longers works :-) I've already cleaned it using a cleaning diskette and
alcohol.
What are my options? Is the floppy drive in a PCjr as goofy/proprietary as the
rest of the machine? If so, should I even attempt to repair it? By the bad
clunk/buzz noise, I am assuming the head is slamming into the side of the drive
or something equally heinous.
Another related question: When I was first getting started with personal
computers 25 years ago, I seem to recall that track alignment was a common
problem and could be fixed by using a calibration diskette and special software
that you could monitor as you turned the alignment screw. Without one of those
factory calibration diskettes, is it even possible to align/calibrate a floppy
drive for track alignment?
--
Jim Leonard (trixter at oldskool.org) http://www.oldskool.org/
Want to help an ambitious games project? http://www.mobygames.com/
Or check out some trippy MindCandy at http://www.mindcandydvd.com/
8 Oct
8 Oct
7:41 p.m.
On Sat, 8 Oct 2005, Jim Leonard wrote:
What are my options? Is the floppy drive in a PCjr as
goofy/proprietary as the
rest of the machine?
The drive is NOT proprietary.
One of the drives that IBM used a lot in them was the Qumetrak 142.
That is the third worst drive I've dealt with.
If so, should I even attempt to repair it?
If it is anything other than the Qumetrak 142.
Another related question: When I was first getting
started with personal
computers 25 years ago, I seem to recall that track alignment was a common
problem and could be fixed by using a calibration diskette and special software
that you could monitor as you turned the alignment screw. Without one of those
factory calibration diskettes, is it even possible to align/calibrate a floppy
drive for track alignment?
Nowhere near as well as you can with a scope and an analog alignment
diskette.
With the digital alignement disk (scopeless), you can do a marginally
adequate job.
With a scope and a diskette that is known to be in alignment (KNOWN TO BE
IN ALIGNMENT, NOT just "from a working machine), you can get the alignment
close enough to work.
9 Oct
9 Oct
5:42 a.m.
As the subject implies, I've got a PCjr with a floppy drive that is having
trouble reading disks and fails the internal CTRL-ALT-INS diagnostics.
Symptoms include a very loud/bad noise at first seek (rest of seeks
sound ok),
This sounds (!) like it's not detecting track 0 correctly, it's just
banging the head into the end stop.
My first thought is to clean the track 0 sensor (has it got dust in the
gap), then sit down with the drive and test gear and debug the circuitry.
and a "B" ("have drive serviced")
result of the diags. The diskette I
used for
diags was tested in a 5150 and found to be good. This drive was
working a week
ago but has declined steadily until it can't successfully boot the
machine. (I
bought some time by formatting disks 8 sectors per track instead of 9,
but that
no longers works :-) I've already cleaned it using a cleaning diskette and
I can't see why that would help, unless the spindle is turning too fast
or something. That would be a very odd problem...
alcohol.
What are my options? Is the floppy drive in a PCjr as
goofy/proprietary as the
No, it's a standard half-height 360K floppy drive. IIRC the original was
a Qumetrack or something, which has a bad reputation. But there is a
schematics in the Options and Adapters TechRef, so it's not going to be
that hard to fix.
rest of the machine? If so, should I even attempt to
repair it? By
the bad
I _would_ have a go. IIRC, it's a fairly easy drive to work on (no ASICs,
mostly common TTL chips), schematics exist, etc. Let me know if you want
to do this (and what test gear you have available), and I'll dig out the
schematic and talk you through it.
clunk/buzz noise, I am assuming the head is slamming
into the side of
the drive
or something equally heinous.
Another related question: When I was first getting started with personal
computers 25 years ago, I seem to recall that track alignment was a common
problem and could be fixed by using a calibration diskette and special
software
that you could monitor as you turned the alignment screw. Without one
of those
factory calibration diskettes, is it even possible to align/calibrate
a floppy
drive for track alignment?
Yes, doing the 'radial alignment' is not that hard. I've never had any
success with the 'digital alignment disks', I prefer a 'catseye disk' and
an oscilloscope conencted to the differential outputs of the read amplifier.
If you have a disk written on a known-good drive (that is, one with
correct alignment), you can in theory look at the read signal in the same
way when reading that disk, finding the points where the read signal
drops to a particular value on each side of the track, and then setting
the head midway between them (this is actually a lot easier than trying
to find the peak level). But it's a lot harder than using the right
alignment disk.
-tony
6:20 p.m.
Tony Duell wrote:
This sounds (!) like it's not detecting track 0
correctly, it's just
banging the head into the end stop.
My first thought is to clean the track 0 sensor (has it got dust in the
gap), then sit down with the drive and test gear and debug the circuitry.
I'm starting from ground zero. Where *is* the track 0 sensor? (a hard thing
to describe without pictures, I know... sorry)
No, it's a standard half-height 360K floppy drive.
IIRC the original was
a Qumetrack or something, which has a bad reputation.
I've had three go out on me in a decade... come to think of it, I've only had 4
floppy drives die on me in a decade, so that indeed makes the Qumetrack the
worst in my experience :-/
--
Jim Leonard (trixter at oldskool.org) http://www.oldskool.org/
Want to help an ambitious games project? http://www.mobygames.com/
Or check out some trippy MindCandy at http://www.mindcandydvd.com/
10 Oct
10 Oct
9:10 a.m.
Tony Duell wrote:
A couple of days ago I got a comment that I was starting at too basic a
level :-). Oh well...
The track 0 sensor detects when the head assemnby is at cylinder 0 -- the
outermost cylinder. Typically it consists of a slotted optoswitch mounted
on the drive chassis and an opaque 'flag' on the head carriage which
interrupts the light beam in the slotted optoswitch when the head is at
cylinder 0.
A slotted optoswtich is typically a black plastic device with a slot in
it. I would think you could find a picture of one on the web somewhere --
try the RS components (not Radio Shack!) or Farnell catalogues.
If this drive is the one I think it is, it was called the 'Slimline
Diskette Srive' by IBM. I have a schematic (although not a mechanical
layout) in front of me.
The track 0 sensor seems to have 4 wires which go to pins A3, A4, B3, B4
of connector J3. Also on this connector are the in-use LED, the spindle
motor and its tachogenerator, and the head stepper motor. You should be
able to find it by tracing wires.
As for the circuitry : A slotted optoswitch consists of an infra-red LED
shining at a phototransistor across the slot. Put something (that 'flag'
I mentioned) in the slot, and the light can't reach the phototransistor.
In this case, the LED is connected between A3 (connected to +5V via a 150
Ohm resistor R61) and B4 (ground). The phototransistor between B3
(ground) and A4 (connected to +5V via a 47k resisotr R43). The last is
the output from the sensor, of course. It also goes to TP8 and to the
input of U16d ('14). The output of that is inverted by U14c ('00, inputs
connected together), then NANDed with a couple of signals from the
stepper motor driver circuit by U7c ('10). THat's inverted again by U16a
('14), then fet to one input of U11b ('38) which drives the track00 pin
on the interface connector (the other input to this gate comes from the
drive-selected signal).
What I would do next is power up the drive as usual, and with the head
stepped away from track 0, look at the output of U14c -- that is pin 8 of
U14 -- with a logic probe. It should be low, and should go high if you
put something opeaque (to IR) in the slot of the sensor.
-tony
This sounds (!) like it's not detecting track
0 correctly, it's just
banging the head into the end stop.
My first thought is to clean the track 0 sensor (has it got dust in the
gap), then sit down with the drive and test gear and debug the circuitry.
I'm starting from ground zero. Where *is* the track 0 sensor? (a hard thing
to describe without pictures, I know... sorry)
7013
days inactive
7014
days old
4 comments
3 participants
participants (3)
-
ard＠p850ug1.demon.co.uk -
cisin＠xenosoft.com -
trixter＠oldskool.org